10-24-2014 12:05 AM
I have two different data sets that I am trying to work with and I need some help! All of the variables have been coded, I've imported the Excel sheets, and now I need to input the code into SAS; however, I'm not too sure what I have to do...
1. The first data set contains: sex, age, area (1, 2 or 3), period (yearly) n, and population. I would like to see if there are differences in rates between the three study areas (and within, if possible) over time. I would like to do this for rates overall, as well as by sex and by age. I think I need to do Poisson regression and have created a logpop variable, but I'm really not sure what to put in for the coding of the actual model.
2. The second dataset is only for one study area and contains: sex, age, period (monthly), n, pop, and well. I would like to do a time series analysis to see if the 'well' variable influences rates over time (overall, as well as by sex and by age). I think I'm looking at PROC ARIMA, but again, I'm not actually sure what to put in for the code.
UPDATE: Actually, I think I have the code for 1 sorted out now. I input 'class period area and age', then to model the rates as a function of study area I just did 'model n= area'. It seemed like the data were overdispersed so I changed from dist=Poisson to dist=NB.
So now I'm just left on figuring out the PROC ARIMA statements